Skip to main content
Fig. 4 | Breast Cancer Research

Fig. 4

From: An integrated deep learning model for the prediction of pathological complete response to neoadjuvant chemotherapy with serial ultrasonography in breast cancer patients: a multicentre, retrospective study

Fig. 4

Visualization and interpretability of SUAS. A The Sankey diagram of changes among the TpCR, TnpCR, FpCR, and FnpCR in difference prognosis factors, with the increment of the predictors in SUAS, all categories (TpCR, FpCR, FnpCR, and TnpCR) presented with significant and constant fluidity. B Entropy, the feature extracted from the ultrasound images of four patients, was visualized in Phases 0–2. It could be observed that even for patients with pCR and NpCR who had the same tumour stage, the similar NAC protocol and age (Patient A vs Patient D, Patient B vs Patient C), the characteristics of the ultrasound images in Phases 0–2 were not apparently different by naked eyes. But after clustering the entropy in the tumour area, it could be clearly found that the clustering of NpCR patients was more scattered than that of pCR patients, with chords and burrs around. Abbreviations: SUAS: serial ultrasonography assessment system; TpCR: true positivity; TnpCR: true negativity; FpCR: false positivity; FnpCR: false negativity; pCR: pathological complete response

Back to article page